DESCRIPTION (Applicant's Abstract): Socio-economic trends and greater longevity in the United States have placed an ever larger number of grandparents in active caregiving roles. Despite such change, little is known about the impact of life histories and health on grand parenting, especially in the adolescent years and in the transition to adulthood. Using longitudinal data on rural generations, this proposed five-year study investigates three interrelated problem foci. The first problem area relates the life experience of adults to their grand parental activities, such as the strength of their ties to grandchildren and their involvement in mentoring. This line of analysis moves beyond a common preoccupation with current intergenerational relations to the important connection between changing lives and relationships. The second problem area is defined by the antecedents and consequences of emotional and physical health among grandparents. The proposed analyses investigate pathways to good health and its association with grand parenting, including the psychosocial benefits of meaningful relations with grandchildren. The third area of inquiry relates variations in grandparent history, health, and behavior to the psychosocial competence of the target children--such as their educational and social success. Activities that bring grandparents and grandchildren together provide a strategic "window of opportunity" for assessing their "linked lives" and competence in terms of quality of interaction, sentiment, and meaning. Data come from members of three generations (children, parents grandparents) in related longitudinal studies, the Iowa Youth and Families Project (IYFP) and the Study of Single Parents (SSP). The IYFP began with 451 two-parent families of 7th grade adolescents in 1989 and includes four waves of data on their perceptions of grandparents and of relations to them. The SSP was launched in 1991 with 207 rural households. The 9th graders were added to the IYFP in the 1994 wave. A fifth wave of data from the grandchildren is proposed for 1999 by using a phone interview (the Study members will be approximately 21). The grandparents were contacted in 1994 with phone interview them again by phone in 1999. Across the three problem areas, the analysis will be tailored to specific questions. Life course and grand parental influences will be assessed with hierarchial linear and structural equation model. Variable-based analyses will be supplemented with person-centered analyses by grouping trajectories and behavior profiles.